Part 3 - Clinical Microsystems

Joint Commission on Quality and SafetyJournal

April 2003 Volume 29 Number 4159

Microsystems in Health Care:Part 3. Planning Patient-CenteredServices

Microsystem Series

Marjorie M. Godfrey, MS, RNEugene C. Nelson, DSc, MPH

John H. Wasson, MDJulie J. Mohr, MSPH, PhD

Paul B. Batalden, MD

On the basis of James Brian Quinn’s originalwork,1 Batalden and Nelson and their col-leagues have made health care adaptations to

the smallest replicable unit concept. This concept, whichwe know as the clinical microsystem—“the small, func-tional, front-line units that provide most health care tomost people”2(p 474)—has become the focus for improve-ment of health care. Based on a study of 20 high-performing clinical microsystems—and on extensivedevelopment and testing in the United States andEurope—a collection of helpful principles, frameworks,and tools have been created to help clinical units takeadvantage of microsystems thinking.

The first article in this series2 introduced the conceptof clinical microsystems as the essential building blocksin all health systems and summarized research results onwhat factors contribute most to high performance inquality and value. The second article3 discussed the useof data and measures to support the work of clinicalmicrosystems. This article shows how microsystems canplan services to best meet the needs of the subpopula-tions of patients that they serve.

In this article we explore the challenge of gainingdeeper knowledge of the “four P’s”—the patients, peo-ple, processes, and patterns—of clinical microsystems.We identify specific activities, information, and knowl-edge that is needed to design and plan patient care andpatient-centered services that meet patient expectationswhile improving the work environment for staff.

The phrase planning patient-centered services refersto the analysis of the inner workings, the architectureand flow—or the “anatomy” and “physiology”—of the

Background: Strategic focus on the clinical microsystems—the small, functional, frontline unitsthat provide most health care to most people—is essen-tial to designing the most efficient, population-basedservices. The starting place for designing or redesign-ing of clinical microsystems is to evaluate the four P’s:the patient subpopulations that are served by themicrosystem, the people who work together in themicrosystem, the processes the microsystem uses toprovide services, and the patterns that characterize themicrosystem’s functioning.

Getting started: Diagnosing and treating a clinicalmicrosystem: Methods and tools have been developedfor microsystem leaders and staff to use to evaluate thefour P’s—to assess their microsystem and design testsof change for improvement and innovation.

Putting it all together: Based on its assessment—or diagnosis—a microsystem can help itself improve thethings that need to be done better. Planning services isdesigned to decrease unnecessary variation, facilitateinformed decision making, promote efficiency by con-tinuously removing waste and rework, create processesand systems that support staff, and design smooth,effective, and safe patient care services that lead tomeasurably improved patient outcomes.

Conclusion: The design of services leads to criticalanalysis of the resources needed for the right personto deliver the right care, in the right way, at the righttime.

Article-at-a-Glance

160April 2003 Volume 29 Number 4

microsystem for the purpose of making services avail-able to best meet the needs of the distinct subpopula-tions served by the practice. In contrast, the phraseplanning patient-centered care refers to the individual-ization of those services (offered by the microsystemitself or by other microsystems in the organization or the community) to best meet the changing needs of individual patients as these people’s conditions, self-management skills, and desires change over time. Byway of analogy, to plan services is to plan the menu forprospective guests whereas, to plan care is to combineand deliver the menu offerings in a manner that meetsthe unique tastes and needs of each individual guest thatrequires service. This article focuses on planning patient-centered services, and the next article will focus onplanning patient-centered care.

Planning Patient-Centered ServicesThe planning of patient-centered services is based onknowledge of (1) the needs of the major subpopulationsof patients served by a clinical microsystem and (2) howthe people in the microsystem interact with one anotherand (3) with their processes to produce critical out-comes. This knowledge comes from both formal analysisand from tacit knowledge of the practice structure, itspatients and its processes, and its daily patterns of workand interaction.

To plan services for their patients, members of themicrosystem benefit by mastering the four P’s:■ Know your patients. Who are we caring for? Arethere subpopulations we could plan services for differ-ently? What are the common patient diagnoses and con-ditions? What other microsystems support what we doto meet patients’ needs? How satisfied are patients withour microsystem?■ Know your people. Who provides patient care, andwho are the people supporting the caregivers? Whatskills and talents do the members need to provide theright service and care at the right time? What is themorale of our team? What is the role of information tech-nology as a “team member”?■ Know your processes. How do we deliver care andservices to meet our patients’ needs? Who does what inour microsystem? Do our hours of operation match theneeds of our patients? What are our core and supporting

processes? How does technology support processes?How do we learn from failure or near misses? ■ Know your patterns. What are the health outcomesof our patients? What are the costs of care? How do weinteract within our microsystem? What does it feel liketo work here? What are the costs of our microsystem?Do information systems provide data and information ina timely way to inform us about the impact of our serv-ices? How do we stay mindful of the possibility of ourefforts failing?

When members of a clinical microsystem worktogether to gain information about their patients, people,processes, and patterns, they acquire knowledge thatcan be used to make long-lasting improvements for theclinical microsystem.

Case Study: Planning Services forSubpopulations of Patients to BestProvide Care for Individual PatientsOne clinical microsystem, Evergreen Woods, a primarycare practice that is part of Norumbega Medical inBangor, Maine, has been evolving for more than a decadeto plan services (in advance) that will be there on demandto provide outstanding care for individual patients.

A Typical VisitWhen a patient calls the office with a medical problem,

a patient representative triages him or her using the TriageCoupler®.* This program is driven by protocols that canhandle a broad spectrum of problems, from the commoncold to complex chest pain, and services, such as stan-dardized protocols for overseas travel and prescriptionrefills. If the patient needs to be seen, prompts are pro-vided for questions to ask, and diagnostic tests that mightbe required before the patient comes to the office are sug-gested. Sometimes, the patient is provided with a standardtreatment and does not need an office visit. This informa-tion tool supports highly trained individuals who do nothave medical degrees to safely and competently makedirect decisions about patient care at the point of contactwith the patient. If a patient ever questions the advicebeing given, an appointment is booked.


* Adapted from Problem-Knowledge Couplers® (PKCs®), available atwww.PKC.com (last accessed Jan 21, 2003).

161

The patient is asked to arrive at the office 30 minutesbefore the appointment to complete a health status survey of his or her physical and mental health status,medical history, presenting problems, and current func-tioning. Medical assistants escort the patient to the examroom, where they use Problem-Knowledge Couplers®

(PKC®)* to verify patient history, to enter current assess-ment data and, when available, to enter test results. Thisinformation allows the physician or nurse practitioner tospend more time with the patient on unique issues,shared medical decision making, and patient education.The provider reviews the findings for possible diagnosesand potential care options. After the appointment’s con-clusion, the patient is given printed information abouthis or her condition, as well as a copy of the visit notethat is stored in the patient’s electronic medical record.

Evergreen Woods collects extensive data on patientdemand for office visits. These data, which include trendcharts by session, day of week, and month of year, allowthe staff to deploy the practice’s resources to matchdemand. Evergreen Woods uses continuous feed forwardinformation and planned feed back systems,3 as well asextensive databases, to aid planning and improvement.The practice staff use a “data wall,” which reports per-formance measures, to monitor progress for the clinicalteam and to identify improvement ideas and actions.4

Evergreen Woods’s pattern of staff interaction includesweekly team meetings, frequent e-mail communications,and many off-hour get-togethers. Continuous training isconducted to train every member on effective interperson-al communication skills. Additional facts that are key ele-ments in the success of Evergreen Woods, along with theirrelationship to the four P’s, are shown in Table 1 (p 162).

Comments Evergreen Woods is an exemplary model of office effi-

ciency and advanced design. Although the practice isheavily invested in technology, most clinicalsettings—be they rural, urban, academic, inpatient, oroutpatient—can learn and adapt the following tips:■ Integrate data into the flow of work to support

the work. Data collection is integrated into the design of

patient care and operations. There are methods and waysto achieve similar results and tracking without advancedinformation technology. Decision making without data isnot acceptable in this clinical environment.■ Enable all staff to make the most of their talent,

training, and skills. Optimization of staff roles,through detailed training and education of each individ-ual, leads to increased abilities to cross-cover for oneanother, engage in improvement work, and feel a senseof accomplishment and self-worth on a daily basis.■ Provide strong leadership. The leader’s example andvision guide the common goals and values of the group.

The next section builds on this case study andexplores methods and tools that can be used to promote“guided discovery” for staff to gain knowledge about thepopulation they serve, the processes for providing serv-ices, and the patterns that (1) spin off the good or badoutcomes for patients and (2) engender a generative ortoxic work environment for staff.

A Developmental Journey: How MightWe Begin to Assess, Understand, andImprove a Clinical Microsystem? Build knowledge of the core processes and outcomes of your microsystem to foster the continual improve-ment and innovation necessary to meet and exceedpatient needs.5

Getting Started: Diagnosing and Treating a ClinicalMicrosystem

Methods and tools have been developed for microsys-tem leaders and staff to use (or adapt to local circum-stances) to assess their microsystems and design tests ofchange for improvement and innovation. The aim is toincrease each microsystem’s capacity to better realize itspotential and to better relate to other critical microsys-tems that come together to form the service continuum.

Every person and microsystem is unique. The toolsand questions found in the Clinical Microsystem Action

Guide6 and Assessing Your Practice Workbook7 and atwww.clinicalmicrosystem.org are intended to provideguidance and to provoke thinking about essential infor-mation that can help to improve a microsystem. Theworkbook provides a framework to diagnose the four P’sof a microsystem, which are now described.



† Problem-Knowledge Couplers® (PKC®) are available at www.PKC.com(last accessed Jan 21, 2003).



Table 1. Evergreen Woods Facts and Links to the Four P’s

163

Know your patients. For 100 years, from Ernest A.Codman to John E. Wennberg, we have known that mostpractitioners have lacked access to data on the vitaldetails of the patient populations they cared for.8,9

Further, they knew even less about distinct sub-populations of patients whose care could be plannedproactively. Key to planning services is knowledge of thesubpopulations served. Table 2 (p 164) offers examplesof some of the variables that clinical microsystemsshould know about the patients they serve.

Gaining this information has been one of the mostpowerful tools for practice members exploring ways toimprove their current delivery system. Thedacare, basedin Kimberly, Wisconsin, learned that approximately 3% ofthe patients in its health plan were diabetic. In partnershipwith the health plan, Thedacare created a registry to trackevidence-based interventions. Group visits evolvedthrough a primary care practice exploring innovative caremodels. The outcomes for patients in the group visits improved significantly. For example, glycosolatedhemoglobin (HgA1C) levels �8 improved 4%, low-density-lipoprotein (LDL) levels �130 mg/dl improved32%, and the overall quality of care being rated by patientsas “excellent” improved 14%.10 The reason for theseimprovements is that the clinical microsystem made aninnovation in service delivery; it moved from sole relianceon one type of service (one-on-one visits with a physician)to offering a second type of service (group visits withstandardized tracking and interventions that are cus-tomized to the individual patient’s needs and wishes) tocomplement the traditional one-on-one visit.

Clinical microsystems need to be acutely aware ofhow patients perceive the care they receive. Whenpatient satisfaction surveys are conducted, the resultsare often sent back to microsystems many months aftera particular set of patients has been seen, which makesit difficult to take timely action to improve services. Abrief, point-of-service patient satisfaction survey, such asthat found in Assessing Your Practice Workbook,7 can beused to provide timely patient-based feedback.

Know your people. Many members of clinicalmicrosystems do not see their own roles and the rolesand functions of others as interdependent, as part of agroup of professionals with an aim and a system to pro-vide care to subpopulations of patients. What are the

morale and level of stress within your clinical setting?What is the turnover rate? Is the right person doing theright thing at the right time for patient care? Is there an appropriate match between function and roles basedon talent, education, training, and licensure? What arethe roles in nonvisit care? The staff in the clinicalmicrosystem make or break the processes of servicedelivery. Without them, good functionality of the clinicalmicrosystem cannot exist. Managing staff as a vitalresource, based on detailed data on patient needs anddemand for services, is essential.

Know your processes. Many health professionalsare “process illiterate.”11 The best way to eliminateprocess illiteracy is flowcharting or process mapping.How much time does it take for patients to receive serv-ices? How much undesirable variation in processesexists? How much waste and rework make the day morefrustrating? How do core and supporting processes getaccomplished? Are they done in the same way by everymember of the team? Are patients assessed in a standardway? Do clinical support staff perform activities thatanticipate the arrival of patients? How does technologysupport work flow and care delivery?

Variation in clinical services is often based on how thephysician wishes to have things done rather than onwhat is the best process for the patient. Table 3 (p 165)provides an assessment tool to help clinical microsys-tems evaluate the services they provide. All staff mem-bers (1) complete the assessment tool, (2) determinewhich process to improve based on the highest-rankedproblem process, and (3) begin to test changes by modi-fying the flowchart of the current process to represent ahoped-for improvement (see Sidebar, p 166).

Know your patterns. The combination of patients,staff, and processes in a particular microsystem results inthe creation of patterns that reflect routine ways of think-ing, feeling, and behaving on the part of both patients andstaff. The patterns are also related to the typical resultsand outcomes—and variations thereof—that are associat-ed with the microsystem’s mission. Some patterns will bewell known and talked about (for example, hours of serv-ice, busy times of the day or week, common hassles, andbottlenecks). Some patterns may be well known and neverdiscussed (sacred cows), whereas some may be unrecog-nized by staff and patients but nevertheless have powerful





Table 2. The Four P’s for Clinical Microsystems Across the Health Care Continuum*

* Note that the variables in the four P’s are similar across the continuum; OR, operating room.

165

effects (mistrust stemming from a local culture dominatedby historical divides that separate staff with different edu-cational backgrounds, such as nurses, receptionists, physi-cians, and technicians). The following questions revealimportant underlying patterns:■ Who is the leader?■ What is the leadership style?■ How do we “act out” the mission of our clinicalmicrosystem every day?■ What are the cultural patterns of norms, sentiments,and beliefs in our practice setting?■ What barriers tend to separate health professionalsand administrative support staff?■ How easy is it to ask a question about patient care?■ How often does the entire staff meet for the purposeof planning services that are patient centered?■ How satisfied are patients with their access to services?■ How do patients feel about the goodness of their out-comes and costs of receiving care?

■ How do we respond to disruptions of our routines?■ How do we “notice” the failure of our systems that wedepend on to prevent accidents and harm to our patients?

Putting It All Together: PlanningServices Based on its assessment—or diagnosis—a microsystem cannow help itself improve the things that need to be done bet-ter. Based on knowledge of the four P’s, what can we proac-tively plan for in our daily work to enhance the functioningof our microsystem? Planning services is designed to■ decrease unnecessary variation, ■ build feed forward and feed back mechanisms forinformed decision making; ■ promote efficiency by continuously removing wasteand rework; ■ create processes and systems that support staff to bethe best they can be; and■ design smooth, effective, and safe patient care servicesthat lead to measurably improved patient outcomes.

Core Processes, Supporting Processes,and “Playbooks”Figure 1 (p 167) offers a panoramic view of a primarycare clinical microsystem. It suggests the interplay ofpatients with practice staff and with processes, which in turn produces patterns that characterize its perform-ance. Typical supporting processes in a primary carepractice include activities such as renewing prescrip-tions, reporting to patients diagnostic test results, andmaking referrals.

Flowcharts can be used to diagram and diagnose each process to learn how to redesign it to maximize efficiency. This is particularly valuable for core or supporting processes whose pattern is to be full of has-sles, bottlenecks, or mistakes. Many clinical microsys-tems have used Assessing Your Practice Workbook7 forguided discovery and for taking actions to redesign theirown services; examples of their work are provided inTable 4 (p 168).

A review of the microsystem improvement efforts towhich we have contributed has uncovered many sourcesof waste that commonly occur. Table 5 (p 169) summa-rizes some of these common sources and provides rec-ommendations to reduce waste and improve efficiency.



■ Process■ Answering phones■ Appointment system■ Messaging■ Scheduling procedures■ Reporting diagnostic test results■ Prescription renewals■ Making referrals■ Pre-authorization for services■ Billing/coding■ Phone advice■ Assignment of patients to your practice■ Orientation of patients to your practice■ New patient workups■ Education for patients/families■ Prevention assessment/activities■ Chronic disease management

* Each of the processes is rated by each staff member on a scale of“works well,” “not a problem,” “small problem,” “big problem,” “totally broken,” “cannot rate,” “we're working on it,” and “source of patient complaint.” If the process is a source of patient complaint,that is noted.

Table 3. Practice Core and SupportingProcesses Assessment*


A clinical microsystem might ultimately build itsown playbook—an organized collection of agreed-on flowcharted processes that is used for training, performance management, and improvement. Theplaybook can be used for educating new staff, cross-training staff, managing performance, and trouble-shooting by providing a reference on how processesshould work.

DiscussionIntentional Planning of Services and the Value ofMeeting for Service PlanningOur study of microsystems in health care revealed thathigh-performing units intentionally designed patient-centered services to support patients and families andthe staff providing care. As shown in the EvergreenWoods case study, planning services is intentional andwell orchestrated. Moreover, it is supported by a contin-uous flow of data (for example, running data throughoutthe day on unfilled slots) to inform every member of themicrosystem, drive corrective actions (any staff personcan schedule patients into unfilled slots anytime duringthe day), and spawn improvements (monthly all-staffmeetings and annual retreats).

The service sector has many examples of people comingtogether to plan the services they deliver. In good restau-rants, waiters, cooks, and hostesses preview the menus forthe day and cover strategies to ensure that the meal serviceis flawless. Plans are made to cover breaks and “what if”scenarios are rehearsed. Flight crews routinely preview theflight plan, use checklists to prepare for takeoff, and reviewflights after their completion because they know all of thiscontributes to a culture of trust and safety.

Similarly, high-performing clinical microsystems havelearned to reap the benefits of daily meetings or huddlesto plan the day and weekly or monthly meetings tostrategize and manage improvement. Holding regularsessions to advance patient-centered care and serviceshas several benefits; it can■ promote collegiality and create an environment ofequality,■ improve communications,■ demonstrate the team of providers to patients andfamilies, and■ help to keep staff members “patient focused.”

Inside-Out PlanningMicrosystems’ overt attention is often focused

more on market-driven service lines or traditionaldepartments, which reflect the strategic plan or budgetsand the organization chart, than on meeting patients’needs through an array of superlative services. Yet focus-ing attention first and foremost on the patient and fami-ly and how they present their health needs to the systemmakes it relatively easy to identify the microsystems thatprovide services and to determine how the best services


Sidebar. Analysis and Improvement of ProcessesA general internal medicine practice analyzed currentprocesses and identified improvements that could leadto better efficiencies and reductions in waste. Everymember of the practice, including the physicians,nurse practitioners, nurses, and secretaries, completedthe Practice Core and Supporting ProcessesAssessment (Table 3, p 165). The assessment revealedthat the diagnostic test reporting process needed tobe improved through shortening of time until report-ing to providers and patients. After flowcharting theprocess, which revealed rework, waste, delay, and longcycle times, the group brainstormed and then rankordered improvement ideas. It decided to test the ideaof holding a “huddle” at the beginning of each day toreview diagnostic test results to determine actions tobe taken. The aim was to eliminate extra phone callsby the patients and delays in action due to waiting forthe provider response. All the group members wouldknow the plan of action after huddling with theprovider.

Using the Plan-Do-Study-Act (PDSA)1,2 format, thegroup conducted its small test of change. Within 2weeks, patient phone calls for laboratory results haddecreased, reflecting the fact that staff were nowcalling patients in a timely manner about their results.

References

1. Nelson EC, et al: Building measurement and data collection intomedical practice. Ann Intern Med 128:460–466, 1998.

2. Langley G, et al: The Improvement Guide: A Practical Approach toEnhancing Organizational Performance. San Francisco: Jossey-Bass,1996.

167



Figure 1. The interplay of patients, people, and processes can result in the best health care for patients and the staff of

the practice.

Core and Supporting Process Diagram for a Primary Care Clinical Microsystem



Table 4. Assessing a Practice’s Discoveries and Actions*

* HgA1C, glycosolated hemoglobin; ADA, American Diabetes Association; URI, upper respiratory infection; RN, registered nurse.

169



Table 5. Common Oversights and Wastes to Consider When Engaging in Improvement Activities*

* RN, registered nurse.


can be designed within each microsystem and how theseservices can be best linked together.

The staff in many microsystems work in a complexenvironment characterized by competing interests, in-efficiencies, hassles, and frustrations due to poorly oper-ating processes. They may feel helpless, that they cannotmake the system work right because the system is run by outsiders. This feeling can be counteracted by working from the inside out, so that staff learn about their patients and the microsystem—and makeimprovements—from the inside out rather than beingtold what to do from the outside in.

Interdependency and InvolvementIt is rare for staff to realize that they are part of a

microsystem that renders identifiable care to subpopula-tions of patients and are fully interdependent with oneanother and patients. The whole of the practice can onlybe as good as the individual components. Staff are oftenso busy trying to do “the job” that they have no time toreflect on the work they do, how they do it, or what theoutcomes of their efforts are. Involvement of all mem-bers of the microsystem is essential to render services.

ConclusionKnowledge of the patients, the people, the processes,and the patterns of a clinical microsystem drives thedesign, redesign, and creation of patient-centered servic-es. The design of services leads to critical analysis of the

resources needed for the right person to deliver the rightcare, in the right way, at the right time. Tools and meth-ods to support the transformation of clinical microsys-tems to yield better results for patients and staff havebeen described and offered for widespread use andadaptation.

The next article in this series will show how amicrosystem can blend together the services it offers toplan care to best meet the needs of each individualpatient.

The authors are grateful to the Robert Wood Johnson Foundation forGrant Number 036103, which supported their research and learningabout clinical microsystems. Special acknowledgement goes to DrCharles Burger and his staff at Evergreen Woods, Norumbega Medical,in Bangor, Maine, for their gracious hospitality and continued collabo-ration for learning. The authors thank Thomas Nolan, PhD, for contin-ued encouragement and support; Coua Early for technical assistance;and Elizabeth Koelsch for manuscript assistance.

J


Marjorie M. Godfrey, MS, RN, is Director, Clinical PracticeImprovement, Dartmouth-Hitchcock Medical Center,Lebanon, New Hampshire. Eugene C. Nelson, DSc, MPH, isDirector, Quality Education, Measurement and Research,Dartmouth-Hitchcock Medical Center. John H. Wasson,MD, is Professor, Community and Family Medicine andMedicine, Dartmouth Medical School, Hanover, NewHampshire. Julie J. Mohr, MSPH, PhD, is Director ofQuality and Safety Research for Pediatrics, University ofChicago, Chicago. Paul B. Batalden, MD, is Director,Health Care Improvement Leadership Development,Dartmouth Medical School. Please address reprint requeststo [email protected].

1. Quinn JB: Intelligent Enterprise: A Knowledge and Service Based

Paradigm for Industry. New York: The Free Press, 1992.2. Nelson EC, et al: Microsystems in health care: Part 1. Learning fromhigh-performing front-line clinical units. Jt Comm J Qual Improv

28:472–493, 2002.3. Nelson EC, et al: Microsystems in health care: Part 2. Creating a richinformation environment. Joint Commission Journal on Quality and

Safety 29:5–15, 2003. 4. Nelson EC, et al: Building measurement and data collection into medical practice. Ann Intern Med 128:460–466, 1998. 5. Nelson EC, et al: Building a quality future. Front Health Serv Manage

15(1):3–32, 1998. 6. Godfrey MM, et al: Clinical Microsystem Action Guide. Hanover,NH: Dartmouth Medical School, 2002.

7. Godfrey MM, Nelson EC, Batalden PB: Assessing Your Practice

Workbook, rev. Dartmouth, NH: Trustees of Dartmouth College, 2002.8. Codman EA: A Study in Hospital Efficiency: As Demonstrated by

the Case Report of the First Five Years of a Private Hospital, reprint-ed. Oakbrook Terrace, IL: Joint Commission on Accreditation ofHealthcare Organizations, 1996.9. Wennberg JE, et al: Hospital use and mortality among Medicare beneficiaries in Boston and New Haven. N Engl J Med 320:1183–1211, 1989.10. Thedacare Diabetes Cooperative Care Storyboard. Presented at TheInstitute for Healthcare Improvement’s Idealized Design of ClinicalOffice Practices Prototype Session 5. Tampa, FL, Apr 9–11, 2000. 11. Batalden PB, Stoltz PK: Fostering the leadership of a continuallyimproving healthcare organization. Qual Lett Healthc Lead 6(6):9–15,1994.

References

Part 3 - Clinical Microsystems

Documents

Transcript of Part 3 - Clinical Microsystems